White mineral oils, called white oils, are colorless, transparent, oily liquids obtained by the refining of crude petroleum feedstocks. In the production of white oils, an appropriate petroleum feedstock is refined to eliminate, as completely as possible, oxygen, nitrogen, and sulfur compounds, reactive hydrocarbons including aromatics, and any other impurity which would prevent use of the resulting white oil in the pharmaceutical or food industry. White oils generally fall into two classes, technical grade and pharmaceutical grade. Technical grade white oils are those suitable for use in cosmetics, textile lubrication, bases for insecticides, and the like. The more highly refined pharmaceutical grade white oils are those suitable for use in drug compositions, foods, and for the lubrication of food handling machinery. The pharmaceutical grade white oils must be chemically inert and substantially without color, odor, or taste. Also, for these applications manufacturers must remove "readily carbonizable substances" (RCS) from the white oil. RCS are impurities which cause the white oil to change color when treated with strong acid. The FDA and white oil manufacturers have stringent standards with respect to RCS which must be met before the white oil can be marketed for use in food or pharmaceutical applications. In particular, 21 C.F.R. .sctn.172.878 (1988) defines white mineral oil as a mixture of liquid hydrocarbons, essentially paraffinic in nature obtained from petroleum and refined to meet the test requirements of the United States Pharmacopoeia XX, pp. 532 (1980) for readily carbonizable substances and for sulfur compounds. The Ultraviolet Absorption Test generally measures the ultraviolet absorbance of an extract in the range of 260-350 nm, which absorbance is then compared with that of a naphthalene standard. This test sets forth limits for the presence of polynuclear compound impurities in the white oil.
White oil must also pass the Hot Acid Carbonizable Substances Test (ASTM D-565) to conform to the standard of quality required for pharmaceutical use. In order to pass this test the oil layer must show no change in color and the acid level is not darker than that of the reference standard colorimetric solution. From this test it will be seen that for purposes of interpreting test results, the art has recognized that a value of 16 or below on the Hellige Amber C Color Wheel is sufficient to pass the carbonizable substances test.
The present invention is primarily concerned with the production of pharmaceutical grade white oils. There are numerous processes in the prior art for the production of white oils of both grades. In general, the first step in the production of white oil is the removal of lighter fractions, such as gasoline, naphtha, kerosene, and gaseous fractions, from the feedstock by fractional distillation. In early processes, white oil was refined by treatment with sulfuric acid to remove unsaturated aromatic and unstable hydroaromatic compounds which comprised most of the impurities present in the oil. Typically, the acid treated oil was subjected to adsorption refining to remove such impurities as carbon, coke, asphaltic substances, coloring matter and the like.
Conventional methods of making white oils with sulfuric acid however, have been subject to objection in recent years since acid treating is costly and gives rise to undesirable amounts of sludge. Because of objections to sulfuric acid treatments, other procedures were developed for the production of white oils from hydrocarbon feedstocks. Representative processes of these procedures can be found in U.S. Pat. Nos. 3,392,112; 3,459,656; 4,055,481; 4,251,347; 4,263,127; and 4,325,804, which are incorporated herein by reference. Further, U.S. Pat. No. 4,786,402, which is also incorporated herein by reference, teaches a two-step catalytic hydrogenation process. The catalyst used in the first step is a sulfur resistant, non-precious metal having both a hydrotreating (especially hydrodesulfurization) and a hydrogenation function. The metal of the catalyst will typically be selected from tin, vanadium, chromium, molybdenum, tungsten, iron, cobalt, nickel and mixtures thereof. The metal is present in a catalytically effective amount, for example, about 2 to 30 weight percent. The catalyst exists as the free metal, or in the form of an oxide or sulfide. Hydrogenation to a medicinal grade white oil is achieved in the second step of the process over a reduced nickel-containing catalyst, which is extremely sensitive to trace sulfur.
Hydrodesulfurization is one of the fundamental processes of the refining and chemical industries. The removal of feed sulfur by conversion to hydrogen sulfide is typically achieved by reaction with hydrogen over non-noble metal sulfides, especially those of Co/Mo and Ni/Mo. The reaction is performed at fairly severe conditions of temperatures and pressures in order to meet product quality specifications, or to supply a desulfurized stream to a subsequent sulfur sensitive process. The latter is a particularly important objective because some processes are carried out over catalysts which are extremely sensitive to poisoning by sulfur. This sulfur sensitivity is sometimes sufficiently acute as to require a substantially sulfur free feed. In other cases environmental considerations and mandates drive product quality specifications to very low sulfur levels.
There is a well established hierarchy in the ease of sulfur removal from the various organosulfur compounds common to refinery and chemical streams. Simple aliphatic, naphthenic, and aromatic mercaptans, sulfides, di- and polysulfides and the like surrender their sulfur more readily than the class of heterocyclic sulfur compounds comprised of thiophene and its higher homologs and analogs. Desulfurization reactivity decreases with increasing molecular structure and complexity within the generic thiophenic class. For example, the simple thiophenes are the more labile, or "easy" sulfur types. The other extreme, which is sometimes referred to as "hard sulfur" or "refractory sulfur," is represented by the derivatives of dibenzothiophene, especially those mono- and di-substituted and condensed ring dibenzothiophenes bearing substituents on the carbons beta to the sulfur atom. These highly refractory sulfur heterocycles resist desulfurization as a consequence of steric inhibition precluding the requisite catalyst-substrate interaction. For this reason, these materials survive traditional desulfurization and they poison subsequent processes whose operability is dependent upon a sulfur sensitive catalyst. Destruction of these "hard sulfur" types can be accomplished under relatively severe high pressure process conditions, but this may prove to be economically undesirable owing to the onset of undesirable side reactions. Also, the level of investment and operating costs required to drive the severe process conditions may be too great for the required sulfur specification.
A recent review (M. J. Girgis and B. C. Gates, Ind. Eng. Chem., 1991, 30, 2021) addresses the fate of various thiophenic types at reaction conditions employed industrially, e.g., 340-425 C. (644-799 F.), 825-2550 psig. The substitution of a methyl group into the 4- position or into the 4- and 6-positions decreases the desulfurization activity by an order of magnitude for dibenzothiophenes. These authors state, "These methyl-substituted dibenzothiophenes are now recognized as the organosulfur compounds that are most slowly converted in the HDS of heavy fossil fuels. One of the challenges for future technology is to find catalysts and processes to desulfurize them."M. Houalla et al., J. Catal., 61, 523 (1980) disclose activity debits of 1 to 10 orders of magnitude for similarly substituted dibenzothiophenes under similar hydrodesulfurization conditions. While the literature addresses methyl substituted dibenzothiophenes, it is apparent that substitution with alkyl substituents greater than methyl, e.g., 4, 6-diethyldibenzothiophene, would intensify the refractory nature of these sulfur compounds. Condensed ring aromatic substituents incorporating the 3,4 and/or 6,7 carbons would exert a similar negative influence. Similar results are described by Lamure-Meille et al., Applied Catalysis A: General, 131, 143, (1995) based on similar substrates.
Mochida in et al., Catalysis Today, 29, 185 (1996) address the deep desulfurization of diesel fuels from the perspective of process and catalyst designs aimed at the conversion of the refractory sulfur types, which "are hardly desulfurized in the conventional HDS process." These authors optimize their process to a product sulfur level of 0.016 wt. %, which reflects the inability of an idealized system to drive the conversion of the most resistant sulfur molecules to extinction. Vasudevan et al., Catalysis Reviews, 38, 161(1996) in a discussion of deep HDS catalysis, report that while Pt and Ir catalysts were initially highly active on refractory sulfur species, both catalysts deactivated with time on oil.
In light of the above, there is still a need for a desulfurization process that can convert feeds bearing the refractory, condensed ring sulfur heterocycles at relatively mild process conditions to products containing substantially no sulfur.